What helps in the short term may hinder later: High grass abundance caused by the sowing of a competitive grass species limits meadow steppic grassland regeneration

With the global loss of grassland ecosystems, biodiversity and ecological stability are declining at an alarming rate. One way to counteract this trend is grassland restoration on abandoned fields. A widely used method is sowing competitive native grasses at high density to rapidly establish a vegetation matrix. Here, we studied the long-term effectiveness of sowing the regionally dominant native grass species, Festuca valesiaca, in an abandoned field in Hungary to restore the region’s natural vegetation, the endangered Pannonic loess meadow steppe. The grass was sown at high density, supplemented by additional seeding of 14 characteristic forbs. We compared vegetation development in this sown site with a neighbouring spontaneously recovering field and an adjacent ancient loess meadow steppe as reference. Vegetation was surveyed annually for 13 years in the restored sites and seven in the ancient grassland. F. valesiaca became dominant four years after sowing, and its cover quickly surpassed that of the ancient grassland. Regeneration was initially rapid but slowed markedly by year six after sowing, showing only small further progress thereafter. In contrast, although weed cover remained high for longer in the spontaneous site, species composition progressed steadily toward the target state, with better establishment of non-sown target forbs. Our findings indicate that a fast-emerging, dense grass cover shortens the weed-dominated phase but slows regeneration later. We highlight the importance of prioritizing long-term continuous monitoring in restoration decisions, as treatments may lead to contrasting outcomes over time.